report resumes - files.eric.ed.gov · report resumes. ed 012 941. predicting a student's...
TRANSCRIPT
REPORT RESUMESED 012 941PREDICTING A STUDENT'S VOCATIONAL CHOICE.BY- HOLLAND, JOHN L. LUTZ: SANDRA W.AMERICAN COLLEGE TESTING PROGRAM, IOWA CITY, IOWAREPORT NUMBER ACT-RR-16 FUB DATEEDRS PRICE MF-$0.25 HC-$0.86 22F.
a.
CG 000 604
MAR 67
_DESCRIPTORS- *PREDICTIVE VALIDITY, *COLLEGE STUDENTS, *CAREERCHOICE, *INTEREST TESTS, CAREER PLANNING, *VOCATIONALCOUNSELING, VOCATIONAL DEVELOPMENT, VOCATIONAL PREFERENCEINVENTORY, AMERICAN COLLEGE TEST FART V
THE PREDICTIVE VALIDITY OF A STUDENT'S EXPRESSEDVOCATIONAL CHOICE WAS COMPARED WITH THE PREDICTIVE VALIDITYOF HIS SCORES ON A VOCATIONAL PREFERENCE INVENTORY. THE DATAFOR THE STUDY WAS FURNISHED BY TWO AMERICAN COLLEGE SURVEYS.STUDENTS FROM TWO NATIONWIDE SAMPLES OF 28 COLLEGES WEREPOLLED FOR THEIR VOCATIONAL CHOICES AND WERE GIVEN THE SIXTHREVISION OF THE VOCATIONAL PREFERENCE INVENTORY. EIGHT MONTHSOR A YEAR LATER, THEY WERE POLLED AGAIN FOR THEIR VOCATIONALCHOICES. VOCATIONAL CHOICES WEVE CATEGORIZED ACCORDING TO ASIX CATEGORY CLASSIFICATION SCHEME WiiICH CODED 99 VOCATIONS -
INTO THESE CLASSES -- REALISTIC, INTELLECTUAL, SOCIAL,
CONVENTIONAL, ENTERPRISING, AND ARTISTIC. RESULTS INDICATETHAT ASKING THE STUDENT ABOUT HIS VOCATIONAL CHOICES ORASKING HIM ABOUT HIS VOCATIONAL INTENTIONS AND ROLE AREALMOST-TWICE AS EFFICIENT AS THE VOCATIONAL_FREFERENCEINVENTORY IN PREDICTING VOCATIONAL CHOICE. THE STUDY SUGGESTSTHAT INTEREST INVENTORIES SHOULD BE USED WITH GREATERDISCRIMINATION. (WR)
PREDICTING A STUDENT'S
VOCATIONAL CHOICE
John L. HollandSandra W. Lutz
Fri RESEARCH
n REPORTSU.S. DEPARTMon ur nutun, tUULItituri 6 WtLfauCC
OFFICE OF EDUCATION
THIS DOCUMENT HA:-, 3IEN REPRODUCED EXACTLY AS RECEIVED FROM THE
PERSON OR ORGANIZATION ORIGINATING IT. POINTS OF VIEW OR OPINIONS
STATED DO NOT NECESSARILY REPRESENT OFFICIAL OFFICE OF EDUCATION
POSITION OR POLICY.
Mar Or 4-90:44 il
Published by Research and Development Division American College Testing Program P. 0. Box 168, Iowa City, Iowa 52240
CG 000 604
Summary
This study compares the forecasting value of a student's initial
report of his vocational choice and vocational role with the Vocational
Preference Inventory. The results of the present study indicate that
we can predict later vocational choices most accurately by one of two
simple methods: (1) ask the student about his first two vocational choices,
or (2) ask him once about his vocational intentions and then ask him for
his preferred vocational role. Either of these methods is almost twice as
efficient as the Vocational Preference Inventory.
Some implications of the findings for the Student Profile Section
of the ACT assessment and for student counseling and research are
discussed.
Predicting a Student's Vocational Choice
John L. Holland and Sandra W. Lutz
For many years, textbook writers, vocational counselors, and
educational researchers have deprecated a student's vocational choice
as being undependable from one year to the next. For sound guidance,
they usually suggest that an interest inventory be used to forecast what
will happen in the student's vocational future. As a result of this belief,
there have been few attempts to learn just how well a student's untutored
vocational choice forecasts his choice at a later date, and how well an
interest inventory predicts this same choice. This comparison is
especially pertinent to the ACT program because, in the Student Profile
Section (Part V of the ACT tests), a student is asked to report his choice
of vocation as well as his choice of major field and his preferred vocational
role. If these items have little forecasting ability, then perhaps an interest
inventory or some other guidance inventory should be substituted.
The goal of the present study is to examine the predictive validity of
a student's choice of vocation and to compare the predictive validity of this
self-expression with his scores on a vocational preference inventory
(Holland, 1965). The present study grew out of some incidental analyses
in an earlier study of college students of superior scholastic achievement
that dem^nstrated the high validity of self-expression and the low validity
of both the Strong Vocational Interest Blank and the Vocational Preference
Inventory (Holland, 1963). The present study repeats this earlier study
but uses a more representative group of college students.
Method
The data for the present study come from two American College
Surveys described earlier by Richards, Holland, and Lutz (1966). Students
were polled for their vocational choices and given the sixth revision of
the Vocational Preference inventory (Holland, 1965). About one year later,
students were polled again for their vocational 'choices.
Students came from two college samples: the fall sample of colleges
included Amherst, Baldwin-Wallace, Cuyahoga Community, California State
at Hayward, Chico State, and the University of Massachusetts. The freshmenin this sample were polled in the fall of 1964 and in May of 1965. The spring
sample of college freshmen was polled in May of 1964 and again in May of
1965, when they were sophomores. The spring sample included the following
colleges: Arkansas Polytechnic, Baylor, Black Hills State, Burlington
Community, California State at Hayward, Colorado State College, Fairmont
State, Indiana State University, Kansas State University, Glassboro State,
Plymouth State, Mount Mercy, Swarthmore, Southeastern State, Southern
Connecticut, Wesleyan, Westbrook Junior, William Jewell, and the Univer-
sities of Alabama, Kentucky, North Dakota and Tennessee. Both samples
contain students with a great range of scholastic potential, vocational interests,
and socio-economic status.
The plan of the study was simple. Student vocational choices were
categorized according to the six-category classification scheme developed
earlier: Realistic, Intellectual., Social, Conventional, Enterprising, and
Artistic (Holland, 1966). Tables 1 and 2 indicate the assignment of vocational
-3-
choices to vocational classes for men and women. Students selected
Table 1A Psychological Classification Scheme for Vocations and Major Fields (Men)
Major Field or Vocation
Agricultural ScienceArchitectureCivil EngineeringFarming
Aeronautical EngineeringAnthropologyAstronomy, AstrophysicsBiochemistryBiologyBotanyChemical EngineeringChemistryDentistryElectrical Engineering
Clinical PsychologyCounseling & GuidanceEducation, General &
Other SpecialtiesEduc. of Excep. ChildrenEducational PsychologyElementary Education
Accounting
EconomicsLawManagementMarketing
ArtArt EducationDramaEnglish, Creative WritingEnglish Education
Realistic ClassForestryGeographyIndustrial Arts Educ.
Intellectual ClassEngineerlg; Gen' 1, OtherEngineering SciencesGeology, GeophysicsMathematics Educ.Math. , StatisticsMedical TechnologyMedicineMetallurgical Eng.Military ServiceNatural Science Educ.
Social ClassExp. & General Psych.Foreign Language Educ.Foreign ServiceGeneral Social SciencesHistoryHistory Education
Conventional ClassBusiness EducationEnterprising Class
Other Business & Comm.Political SciencePublic Administration
Artistic ClassGeneral HumanitiesJournalism, Radio-TV,
CommunicationLiteratureMusic
Industrial EngineeringMechanical EngineeringTrade & Industrial Educ.
OceanographyOther Biolog. Sci. FieldsOther Health FieldsPharmacyPhysical TherapyPhysicsPhysiologyVeterinary ScienceZoology
Ind. & Personnel Psych.Physical Educ. ,
Recreation 8c, HealthSocial WorkSociologyTheology, Religion
Finance
Public RelationsPurchasingSales 1.
Music EducationOther Fine &
Applied ArtsPhilosophySpeech
Table 2
A Psychological Classification Scheme forVocations and Major Fields (Women)
Major Field or Vocation
Agricultural ScienceArchitectureBiochemistryBiology
Clinical PsychologyDentistryExp. & General Psych.Mathematics Educ.
Business Education
Educational Psych.Management, Bus. Ad.
Counseling & GuidanceEduc. , General &
Other SpecialtiesEducation of
Exceptional ChildrenElementary EducationEnglish EducationHistory
Accounting
Marketing
ArtArt EducationDramaEnglish, Creative WritingForeign Service
Intellectual ClassChemistryMath. , StatisticsMedicineNatural Science Educ.
Social-Intellectual ClassMedical TechnologyNursingOther Health FieldsPharmacy
Social-Conventional ClassClerical, Office Work
Social-Enterprising ClassPurchasing
Social-Artistic ClassHistory EducationHome EconomicsHome Economics Educ.HousewifeLawModern Foreign
Language Education
Conventional Class
Enterprising Class
Artistic ClassJournalism, Radio-TV,
CommunicationLibrary Science,Archival Science
Literature
Other Biol. SciencesPhysicsVeterinary MedicineZoology
Physical TherapyPolitical Science, Govt. ,
International RelationsTheology, Religion
Secretarial Science
Sales
Physical Educ. ,
Recreation & HealthPublic Rel. , AdvertisingSocial ScienceSocial Work, Group WorkSociologySpeech
Modern Foreign LanguageMusicMusic EducationOther Fine & Applied ArtsPhilosophy
-5-
their career choices from a coded list of 99 careers. All classifications,
as well as scoring and then establishing the highest Vocational Preference
Inventory Scale Score, were performed by a computer with, we assume,
perfect reliability. Tables were then formed to show how students' first
vocational choices are related to their second or final choices. The time
intervals between choices were 12 months for the spring sample and 8
months for the fall, Using a student's highest scale score, we examined
the predictive validity of the Vocational Preference Inventory in the same
way.
Vocational Choice and VPI
The prediction of a student's final vocational choice from his first
choice eight months earlier is shown in Table 3 for men in the fall sample.
Table 3Prediction of Final Vocational Choice from First Vocational Choice
(Fall Sample, Men, N=1359)
1stVocational Choice
Final Vocational Choice
Real Int Soc Cony Ent Art Und HitsNoRes
Realistic* 106 13 4 1 12 14 67.9 6
Intellectual 31 355 21 14 29 8 57 67.6 10
Social 3 8 110 1 12 3 19 68.8 4
Conventional 1 2 42 7 7 71.2
Enterprising 4 5 15 10 155 4 32 67.1 6,
Artistic 1 2 8 6 48 10 63.2 1
Undecided 4 14 13 4 18 4 53 1
No Response 10 8 5 8 9 1
N
156
525
160
59
231
76
1.1.1
41
The percentage of correct predictions varies from 63.2 to 71.2 percent.
The total number cf correct predictions always exceeds base rate expecta-
tions and cannot be attributed to chance.
The prediction of a student's final vocational choice from his highest
score (High Point Code) among six Vocational Preference Inventory scores,
obtained eight months earlier, is given in Table 4. The percentage of correct
predictions ranges only from 21.5 to 51.4 percent. In this instance, simply
asking the student is clearly superior to using the Vocational Preference
Inventory.Table 4
Prediction of Final Vocational Choice from VPI High-Point Code(Fall Sample, Men, N=1359)
VPIHigh-Point Code
Final Vocational Choice
Real Int Soc Cony Ent Art Und
.
%
HitsNoRes N
Realistic 43 28 9 7 8 17 37.7 2 114
Intellectual 70 282 45_ 9 43 14 70 51.4 16 549
Social 8 32 78 9 34 11 37 36.8 3 212
Conventional 11 19 4 33 23 13 31.4 2 105
Enterprising 9 16 17 18 98 2 29 50.8 4 193
Artistic 8 31 28 1 41 40 35 21.5 2 186elNote. --To make single predictions from the VPI, it was necessary to omit
students whose two highest scores were tied. This occurrence then necessitatedthe omission of students with tied profiles from the tables of expressed choiceso that the comparisons of the VPI and expressed choice are based on identicalsamples. If, however, "expressed" choice predictions are based on all students(with and without VPI ties), the differences in predictive efficiency shown inTables 3, 6, and 9 vary only 1 per cent or less.
The discrepancy between the efficiency of a student's expressed
choice and his VPI scores becomes even greater when we select a sub-
sample of students whose first two vocational choices, at the time of
initial testing, fall in the same vocational class: for example, physics
and chemistry, education and social work, art and literature, etc. The
exact questionnaire items for this analysis were as follows:
My present career choice is: (Select the appropriate number fromthe list of coded careers and curricula)
If I could not have my first choice (above) I would select the followingoccupation: (Select the appropriate number from this list)
Table 5 reveals that sub-grouping students whose first two choices
belong to the same vocational class results in a substantial gain in predictive
efficiency. Correct predictions range from 73.3 to 85.7 percent as contrasted
with 63.2 to 71.2 for the total sample of men. The subgroup of students whose
Table =5
Prediction of Final Vocational Choice for Students Whose FirstTwo Choices Fall in the Same Class
(Fall Sample, Men, N=586)
1st & 2ndVocational Choices
Final Vocational Choice
Real Int Soc Cony Ent Art Und HitsNoRes N
Realistic 36 2 1 6 1 76.6 1 47
Intellectual 18 239 5 4 10 4 21 78.1 5 306
Social 47 3 1 9 78.3 60
Conventional 12 2 85.7 14
Enterprising 1 2 2 69 2 9 78.4 3 88
Artistic 3 3 22 1 73.3 1 30
Undecided 4 1 2 1 5 13
No Response 9 3 3 6 7 28
first two choices fall in different vocational classes are, as expected, less
predictable than students whose first two choices belong to the same class.
Although the percentages of hits are lower--53. 0 to 66. 7--they still exceed
the percentages obtained by the use of the VPI--21. 5 to 51.4.
The results for women in the fall sample are presented in Tables 6,
7, and 8. The main findings for women parallel those for men. The
analyses for women are identical to those for men except for the use of
a special classification scheme developed for women in which the Social
Class is divided into four sub-classes. Expressed vocational choices
predict 34.8 to 83. 8 percent of later choices (Table 6).
On the average, the VPI in Table 7 predicts final choices less
Table 6
Prediction of Final Vocational Choice from First Vocation_ al Choice(Fall Sample, Women, N=1386)
Final Vocational Choice1st Soc Soc Soc Soc %Vocational Choice Int Int Cony Ent Art Cony Ent Art Und Hits NR TotalIntellectual 54 19 2 2 18 2 4 15 42. 5 11 127
Social Intellectual 10 183 3 6 42 1 1 10 7 66.3 13 276
Social Conventional 17 1 1 3 73. 9 1 23
Social Enterprising 2 8 10 1 1 34. 8 1 23
Social Artistic 4 20 4 4 553 1 25 37 83.8 12 660
Conventional 2 1 1 5 55. 6 9
Artistic 1 3 2 34 73 19 52.1 8 140
Undecided 1 7 1 1 28 11 24 12 85
No Response 5 4 16 2 7 9 43
Table 7
Prediction of Final Vocational Choice from VPI High-Point Code
(Fall Sample, Women, N=1386)
VPIHigh-Point Code
Final Vocational Choice
IntScc SocInt Cony
SocEnt
SocArt Cony Ent Art Und
%Hits
Realistic 2 10.0
Intellectual 46 64 4 2 55 1 17 25 20.3
Social 14 111 18 7 481 3 1 38 54 81.9
Conventional 5 9 5 2 12 5 3 5 10.0
Enterprising 7 3 16 1 _ 2 5 0.0
Artistic 12 43 4 7 139 1 1 65 23 20.5
NR Total3
13 227
26 753
4 50
2 36
22 317
Table 8Prediction of Final Vocational Choice for Students Whose First
Two Choices Fall in the Same Class(Fall Sample, Women, N=545)
1st & 2ndVocational Choice
Final Vocational Choice
IntSoc SocInt Cony
SocEnt
SocArt Cony Art Und
Intellectual 25 7 1 1 3
Social Intellectual 2 85 2 2 5 1 1
Social Conventional 8 1 1
Social Enterprising 1 1
Social Artistic 2 8 1 299 13
Conventional 1
Artistic 1 1 8 23 2
Undecided1
No Response l 3 5 5
To
Hits NR Total65.8 1 38
85.0 2 100
72.7 1 11
50.0 2
88.5 7 338
1
60.5 3' 38
1
1 16
=-10-
efficiently than asking the student (0. 0 to 81.9 percent). And the forma-
tion of a sub-sample of women whose first two choices belong to the same
class yields the most efficient predictions--50. 0 to 88. 5 percent. Table8 again demonstrates that students whose first two choices belong to the
same vocational class are more predictable than students whose choices
belong to different classes (32. 6 to 78. 9 percent).
Because of the controversial character of the results obtained for
the fall sample, the same analyses were performed for the spring sample
with similar results. These analyses involve a longer interval of time
(one year instead of eight months) and a different stage of college life
(the end of the freshman year to the end of the sophomore year, asopposed to the beginning of the freshman year to the end of the freshman
year). Table 9 is a summary of the main results. Without exception,
the results in Table 9 replicate what we found earlierexpressed voca-
tional choice is clearly and substantially superior to the Vocational Pre-
ference Inventory.
Table 9
Summary for Spring Sample
Kinds of Prediction
% Correct PredictionsMen
(N=1773)Women(N=2336)
Expressed Vocational Choice--Total 68.7 78. 2
VPI-- Highest Scale 45. 1 59.6
Expressed Vocational Choice--Same 82. 5 86.4
Expressed Vocational Choice--Different 64. 2 71.9
3rWm4"1:4--7,-
-11-
Vocational Choice and Role
The following analyses were performed to see how well we couldpredict a student's later vocational choice when his preferred vocationalrole was considered along with his first vocational choice or his VPI scores.When he took the American College Survey, he responded to the followingitem:
What special role would you like to play within your present occupationalchoice? (Mark one)Being a practitioner of my occupationTraining or teaching others about my occupationLeading or supervising peopleDoing research in my fieldActing as a consultant or expert to othersUndecidedOther role
Table 10, a summary of 24 tables, presents the predictive efficiencies ofa student's highest VPI score and his role preference, and his expressed voca-tional choice and his role preference. With one exception, students who preferthe roles- of practitioner, teacher, leader, researcher, or consultant are morelikely to give the same or closely related vocational choice eight months laterthan are students who fail to respond, or who respond "undecided" or "otherrole. " The underlined percentages within the six classes indicate the classesand vocational roles that appear to go together--where the best predictionsshould be obtained. The efficiency of the predictions obtained for expressedchoice and vocational role approximate those obtained by sub-grouping studentswhose first two choices belong to the same vocational class.
In short, we can predict vocational choices most accurately by one oftwo simple methods: (1) ask the student about his first two vocational
-12-
Table 10Predicting Vocational Choice from First Two Vocational Choices
and Preferred Vocational Role (Fall Sample)
Voc'l Preferred Vocational Role, Men (N=1207)Choice Uth., Oth., NR Pract. Teacher Leader Research. Consult.
f % hits f % hits f % hits f % hits f % hits f % hits1st & 2ndSame
Real 8 75.0 18 83.3 1 0.0 4 75.0 9 66.7 7 85.7Int 54 70.4 94 84.0 18 83, 3 9 33.3 120 80.8 11 63. 6Soc 11 54.5 25 80.0 11 90.9 5 80.0 4 100.0 4 75.0Cony 3, 66.7 5 100.0 1 100.0 0 0.0 1 100.0 4 75.0Ent 19 78.9 35 85.7 1 100.0 21 85.7 2 0.0 10 50.0Art 4 50.0 20 70.0 1 100.0 3 100.0 1 100.0 1 100.0Total 69. 7 82. 7 84. 9 73. 8 79. 6 67. 6
1st & 2ndDifferent
Real 31 58. 1 42 66. 7 6 66. 7 12 75. 0 12 50. 0 6 83. 3Int 75 48. 0 77 55. 8 19 47. 4 13 46. 2 29 58. 6 6 83. 3Soc 25 52.0 32 75. 0 13 84.6 14 50. 0 9 22. 2 7 85. 7Cony 17 47.1 12 75.0 0 0.0 2 0.0 1 100.0 13 92. 3Ent 47 51. 1 52 63. 5 3 66. 7 22 68. 2 5 60.0 14 64. 3Art 18 50. 0 14 57. 10 70. 0 1 100. 0 2 50. 0 1 O. 0Total 50.7 63. 3 64.7 59.4 51. 7 78. 7
Women (N=1258)1st & 2ndSame
hit 7 42.9 12 83.3 2 50.0 17 64.7Soc-Int 22 95. 5 55 81.8 7 71.4 5 80.0 11 90. 9Soc-Cony 2 50. 0 5 80. 8 3 100. 0 1 0. 0Soc-Ent 1 100.0 1 0.0Soc-Art48 79.2 194 92. 3 41 80. 5 29 89. 7 5 60. 0 21 95. 2Cony 1 0. 0EntArt 9 55.6 17 64.7 8 75.0 1 100. 0 3 O. 0Total 76. 4 88. 0 77. 4 88. 6 72. 7 80. 0
1st & 2ndDifferent
Int 27 14. 8 26 53. 8 3 0, 0 1 100. 0 27 25. 9 5 60. 0Soc-Int 40 42. 5 74 68. 9 19 36.8 24 50.9 13 61. 5 6 50. 0Soc-Cony 2 50.0 8 75.0 1 100.0 1 100. 0Soc-Ent 1 0.0 10 30. 0 2 0. 0 4 75. 0 4 25. 0Soc-Art 93 74.2 140 84.3 34 73. 5 28 82.1 11 63.6 16 75. 0Cony 6 66.7 1 100.0 1 0.0EntArt 30 30. 0 42 61.9 13 38.5 4 75.0 2 0.0 11 63.6Total 52. 3 72. 8 52. 8 67. 7 41. 5 62. 8
-13-
choices, or (2) ask him once about his vocational intentions and then ask himfor his preferred vocational role. Either of these methods is almost twice asefficient as the Vocational Preference Inventory.
Discussion
Most of all, the results suggest that educators, researchers, and coun-
selors should make greater use of a person's expressed vocational choices and
that interest inventories should be used with more discrimination. The resultsalso raise a number of questions: (1) Would we obtain similar resultsthesuperiority of expressed over measured interests--if we had used well
established inventories like the Strong Vocational Interest Blank or the Kuder
Preference Record? (2) Does the classification scheme used in the presentstudy provide the kinds of predictions students and counselors want, and isit useful for this purpose? And (3) would we obtain similar differential validi-ties over longer intervals of time and for people of different ages ?
The data for satisfactory answers to these questions are either not
available or not sufficient. For example, the earlier study revealed that six
scales of the Strong were not as efficient as expressed choice for similar
samples1 over a four-year interval. The percentages of hits (prediction
over four years to the same six-category system) equaled 28.2 percent
for the Strong and 56.3 percent for expressed choice (see Tables 4 and 8,
Holland, 1963). This experience suggests, but does not demonstrate, that
counselors using all scales of the Strong or some other interest inventory
would not be expected to surpass expressed choice in other predictive studies.
1The two samples differ because 74 rather than 100 percent of thestudents filled out the Strong.
-14-
We need comparative studies of expressed and measured interests
employing a single classification scheme on the same population for several
instruments and for several time intervals. A review of the predictive
validities for the Strong and the Kuder inventories versus expressed choice
quickly reveals a morass of criteria, classifications, predictive formulae, and
counseling contaminations that usually defy any reliable extrication (Berdie,
1950; Strong, 1943; Dar ley & Hagenah, 1955). This situation prevails
because most predictive studies are primarily concerned with establishing
the validity of an inventory so that the predictive validity of expressed choice,
if studied at all, receives only cursory treatment.
The predictive efficiency of expressed choice in the present study
is due mainly to the use of a classification system. The present classifica-
tion, like many classification schemes, groups similar vocational choices
so that it is possible to distinguish small and large differences. Such an
orientationcalling attention to occupational groups rather than single
occupations--is precisely what students need and what counselors strive
for Because we have usually evaluated the predictive efficiency of
expressed choice in terms of identical vocational choices from one time
to the next, we have treated any change in vocational choice as a gross
change and missed the opportunity to examine the different degrees of
change that a useful classification reveals.
If the present study and its predecessor are persuasive, then we
could abandon the routine use of interest inventories in freshman orienta-
tion programs and rely on what students tell us. Those students who are
undecided or who give successive choices that fall in different vocational
-15-
classes might be given the option of taking an into rest inventory. In
making predictions in vocational counseling, it may be constructive to
rely more upon a person's vocational choice and history of such choices
than upon interest inventories. Interest inventories may be most useful
for characterizing the poles of a person's conflicts about vocations and
for similar diagnostic and treatment purposes. The potential values of
this orientation need more investigation, but it seems unwise to continue
to believe that interest inventories are always needed in the sense that
one always needs a yearly physical.
In the ACT program, the Student Profile Section items on vocational
choice and vocational role clearly have substantial forecasting ability.
And until interest inventories approach the same high level of efficiency,
there is no compelling evidence for the introduction of an interest inventory
into the ACT assessment. Colleges which want to do so can identify students
in need of counseling by finding "undecided" 'students via the Student Profile
Section, or students who cannot designate a vocational role they prefer
to play.
Looking ahead, it appears quite plausible that interest inventories will
have greater use as vehicles for creating better classification schemes for
occupations, preferences, client problems, and occupational material, and
for theoretical work generally. The great majority of counseling problems
may become amenable to various classifications of expressed choices and
associated theories.
References
Berdie, R. F. Scores on the Strong Vocational Interest Blank and the
Kuder Preference Record in relation t self ratings. Journal of
Applied Psychology, 1950, 34, 42-49.
Dar ley, J. G. , & Hagenah, Theda. Vocational interest measurement.
Minneapolis: University of Minnesota Press, 1955.
Holland, J. L. Explorations of a theory of vocational choice and achievement:
IL A four-year prediction study. Psychological Reports, 1963, 12,
547-594.
Holland, J. L. Manual for the Vocational Preference Inventory. (6th
rev. ). Iowa City, Iowa: Educational Research Associates, 1965.
Holland, J. L. A psychological classification scheme for vocations and
major fields. Journal of Counseling Psychology, 1966, 13, 278-288.
Richards, J. M. , Jr. , Holland, J. L. , & Lutz, Sandra W. The prediction
of student accomplishment in college. ACT Research Report No. 13.
Iowa City, Iowa: American College Testing Program, 1966.
Strong, E. K. , Jr. Vocational interests of men and women. Stanford, Calif. :
Stanford University Press, 1943.
ACT Research Reports
This report is the eighteenth in a series published by the Researchand Development Division of American College Testing Program. Reportsare published monthly and mailed free of charge to educators and otherinterested persons who have asked to be on the special research reportmailing list.
The research reports have been deposited with the American Docu-mentation Institute, ADI Auxiliary Publications Project, PhotoduplicationService, Library of Congress, Washington 25, D. C. (ADI Documentnumbers and prices are given below. ) Printed copies may be obtainedfree from the Research and Development Division, American CollegeTesting Program. Photocopies and 35 mm. microfilms may be obtainedat cost from ADI by citing ADI Document number. Advance payment isrequired. Make checks or money orders payable to: Chief, Photo-duplication Service, Library of Congress.
Reports in the series are:
*No. 1 A Description of American College Freshmen, by C. Abe,J. L. Holland, Sandra W. Lutz, & J. M. Richards, Jr.(ADI Doc. 8554; photo, $8. 75; microfilm, $3. 00)
*No. 2 Academic and Non-academic Accomplishment: Correlatedor Uncorrelated? by J. L. Holland, & J. M. Richards, Jr.(ADI Doc. 8555; photo, $3. 75; microfilm, $2. 00)
No. 3 A Description of College Freshmen: L Students with Dif-ferent Choices of Major Field, by C. Abe. , & J. L. Holland(ADI Doc. 8556; photo, $7. 50; microfilm, $2. 75)
*No. 4 A Description of College Freshman: IL Students with Dif-ferent Vocational Choices by C. Abe. , & J. L. Holland(ADI Doc. 8557; photo, $7. 50; microfilm, $2. 75)
*No. 5 A Description of Junior Colleges, by J. M. Richards, Jr. ,
Lorraine M. Rand, & L. P. RandADI Doc. 8558; photo, $3. 75; microfilm, $2. 00)
*No. 6 Comparative Predictive Validities of the American CollegeTests and Two Other Scholastic Aptitude Tests, by L. Munday(ADI Doc. 8559; photo, $2. 50; microfilm, $1. 75)
*This report now available only from ADL
ACT Research Reports (cont. )
No. 7 The Relationship Between College Grades and AdultAchievement. A Review of the Literature, by D. P. Hoyt(ADI Doc. 8632; photo, $7. 50; microfilm, $2. 75)
No. 8 A Factor Analysis of Student "Explanations" of TheirChoice of a College, by J. M. Richards, Jr. , & J. L. Holland(ADI Doc. 8633; photo, $3. 75; microfilm, $2. 00)
No. 9 Regional Differences in Junior Colleges, by J. M. Richards,Jr. , L. P. Rand, & Lorraine M. Rand(ADI Doc. 8743; photo, $2. 50; microfilm, $1. 75)
No. 10 Academic Description and Prediction in Junior Colleges, byD. P. Hoyt, & L. MundayiADI Doc. 8856; photo, $3. 75; microfilm, $2. 00)
No. 11 The Assessment of Student Accomplishment in College, byJ. M. Richards, Jr. , J. L. Holland, & Sandra W. Lutz(ADI Doc. 8955; photo, $3. 75; microfilm, $2. 00)
No. 12 Academic and Non-academic Accomplishment in a Repre-sentative Sample taken from a Population of 612, 000, byJ. L. Holland, & J. M. Richards, Jr.(ADI Doc. 8992; photo, $3. 75; microfilm, $2. 00)
No. 13 The Prediction of Student Accomplishment in College, byJ. M. Richards, Jr. , J. L. Holland, & Sandra W. Lutz(ADI Doc. 9020; photo, $5. 00; microfilm, $2. 25)
No. 14 Changes in Self-Ratings and Life Goals Among Studentsat Colleges with Different Characteristics, by R. W. Skager,J. L. Holland, & L. A. Braskamp(ADI Doc. 9069; photo, $3. 75; microfilm, $2. 00)
No. 15 Can Computers Write College Admissions Test? , by S. M.Richards, Jr.(ADI Doc. 9174; photo, $2. 50; microfilm, $1. 75)
No. 16 Changes in Self-Ratings and Life Goals as Related toStudent Accomplishment in College, by R. W. Skager,& L. A. Bra skamp(ADI Doc. 9214; photo, $2. 50; microfilm, $1. 75)
No. 17 Family Income and the Characteristics of College-BoundStudents, by L. L. Baird(ADI Doc. number not yet available)